Researchers in Canada have used Illumina’s Genome Analyzer to profile microRNA expression in human embryonic stem cells and to discover variants of known miRNAs as well as more than 100 novel microRNA genes.
Their study, which was published online in Genome Research last week, is not the first using Illumina’s sequencing platform to sample small RNAs. But it showcases the usefulness of the technology, and potentially others, for characterizing small RNA in depth.
The researchers, from the BC Cancer Agency in Vancouver, BC, sequenced a small RNA library from pluripotent human embryonic stem cells and one from differentiated cells derived from these stem cells.
Each library was sequenced on a single lane of an Illumina GA flow cell and yielded approximately 6 million unfiltered sequence reads. The researchers mapped a filtered set of these to the human reference genome, 750,000 unique small RNA sequences in total. They then annotated these sequences, based on known small RNA genes such as miRNAs, rRNAs, and tRNAs, and identified 334 known miRNA genes.
However, around 40,000 small RNA sequences in each library could not be annotated. In order to further characterize them, the scientists used both publicly available and in-house algorithms to identify candidates for novel miRNAs and found 83 unique miRNA sequences from as many as 104 genes.
They also found that many of the known miRNAs occurred in different isomeric variants, so-called isomiRs, which might increase the target repertoire of the miRNA gene they derive from.
Finally, they found that almost 145 of the profiled microRNAs appeared to be differentially expressed between the two cell types.
Because of the large number of sequence reads that Illumina’s technology generates in a fraction of a run, it allowed the researchers to sample a wider range of miRNA expression levels, at a lower cost, than others have done using 454’s sequencing technology (see In Sequence’s sister publication, GenomeWeb Daily News 6/26/2006).
While 454’s platform covers about four orders of magnitude of miRNA expression levels in a run, Illumina’s sequencer provides six orders of magnitude per run, according to Ryan Morin, a research assistant in Marco Marra’s group at the BC Cancer Agency and the lead author of the study. This allowed the researchers to analyze not only highly expressed microRNAs, but also low-abundance ones represented by only a few sequence reads.
“We do know that certain cloning procedures show certain biases, and thus cloning frequencies do not necessarily reflect actual abundance of molecules in the cell in all cases.”
“For sure, the technology gave us an edge,” Morin told In Sequence last week.
However, he added that 454’s sequencing platform would be more suitable than Illumina’s for sequencing many different samples in parallel because it allows users to attach barcodes to the DNA. Because of the short read length, that is currently not feasible with Illumina’s platform, he said.
“I agree that Illumina has advantages over 454 as far as sequencing of small RNAs is concerned,” said Eugene Berezikov, a group leader at the Hubrecht Institute in Utrecht, the Netherlands. “This is simply due to the substantially larger number of reads generated by [Illumina] for the same price,” he said.
He added his group has experience with 454’s, Illumina’s, and ABI’s next-generation sequencing platforms, and said that the latter might also be well suited for the task but that “there is still not enough data available to judge the performance of SOLiD.”
Berezikov is also a co-founder of InteRNA Genomics, a startup company that provides bioinformatics services for “deep sequencing data,” in particular for small-RNA analysis.
Though sequencing is his preferred method for miRNA profiling, Berezikov cautioned that library preparation can bias the method. “We do know that certain cloning procedures show certain biases, and thus cloning frequencies do not necessarily reflect actual abundance of molecules in the cell in all cases,” he said.
Other groups have also used Illumina’s system to analyze small RNAs. For example, a group led by researchers at the Broad Institute published a study on microRNAs in Drosophila in Genome Research last year in which they generated almost 800,000 reads on Illuimna’s platform from libraries from Drosophila ovaries and testes.
Also last year, researchers led by Greg Hannon at Cold Spring Harbor Laboratory published a study in Genes & Development in which they sequenced small RNA libraries from mouse mammary epithelial cells using Illumina’s sequencer.
As part of a study published last month in Genes & Development by a research group led by the Sloan-Kettering Institute, scientists sequenced small RNA prepared from Drosophila heads, using a single lane of an Illumina GA flow cell.
In addition, scientists from Tom Tuschl’s lab at Rockefeller University published a protocol] for preparing cDNA libraries from small RNAs in Methods last month that includes the use of 454’s and Illumina’s sequencing technology.
But according to Morin, what is new in his study is the methods he developed, as part of his thesis project, for handling large amounts of sequence data and for discovering and characterizing miRNAs. He plans to make the software, called myRNA, which contains publicly available tools as well as tools he wrote, available once he has made it “sufficiently portable and generally applicable.”
Morin said that he is currently working on new applications for the Genome Analyzer, such as transcriptome sequencing. These applications might also include more clinical ones, such as profiling microRNAs in cancer, he said.
Marra’s group, which currently has five Illumina Genome Analyzers installed, is an early-access customer of the technology that obtained one of the first Solexa instruments back in 2006.